See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/242630869

A review of sulfate removal options for mine waters

Article · January 2004

CITATIONS READS

25 1,161

1 author:

Rob Bowell
SRK Consulting
105 PUBLICATIONS   2,005 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Geochemistry of mine waste View project

Rocks and Minerals View project

All content following this page was uploaded by Rob Bowell on 04 August 2014.

The user has requested enhancement of the downloaded file.

A REVIEW OF SULFATE REMOVAL OPTIONS
FOR MINE WATERS

R.J.Bowell

SRK Consulting, Windsor Court, 1 Windsor Place, Cardiff CF10 3BX,

Wales

Abstract
Sulfide oxidation is a common phenomenon associated with many
mined ore bodies and extracted or processed waste. Where the products of
these chemical reactions can enter a water body the result is often a reduc-
tion in water quality through the increase of acidity, metals and dissolved
salts. Due to an attributed lower environmental impact the release of sul-
fate has received little attention in many regulatory jurisdictions when
compared to control of dissolved metals or acidity. Consequently the lit-
erature on sulfate removal from mine waters is comparatively small when
compared to metal control despite the use and development of several
technologies to reduce sulfate and total salt loadings. Sulfate control lev-
els are based primarily on the secondary drinking water recommendations
of approximately 500 mg/L, based on the laxative effect of high magne-
sium sulfate concentration.
Sulfate control in mine waters primarily follows one of two methodolo-
gies; (1) Removal through membrane separation of salts from water; (2)
Removal of sulfate by salt precipitation through ion exchange, permeable
reactive barrier, biological reduction or formation of insoluble mineral pre-
cipitate.
These approaches are reviewed within this paper and an economic and
technical comparison made between the available technologies. Based on
demonstrated technology and economic benefits the most promising tech-
nologies to date are biological sulfate reduction, SAVMIN, and GYPCIX.
At sulfate concentrations less than 2000 mg/L the limestone/lime process
is an effective low-cost or pre-treatment removal option for sulfate. As
with any mine water treatment option site-specific conditions will control
the most suitable option for a particular mining operation.
2 R.J.Bowell

1. Introduction
Acid rock drainage mechanisms involve the oxidation of sulfide miner-
als and can lead to highly acidic, metal-rich waters with high sulfate con-
tent. However sulfate has a potential corrosive and purgative effect
(AWWA, 1999). Waters rich in sulfate also have a high scaling potential.
In South Africa, for example, it is estimated that 75% of gold mines have
scaling processes essentially related to saturation of water with respect to
CaSO4 (Juby, 1989). Increasingly sulfate is being considered as one of the
more significant long term water quality issues for mining operations and
process plants, particularly in countries with problems of fresh water sup-
ply such as South Africa and Australia (Bosman, 1985; Maree et al.,
1989; Adlem et al., 1991; Du Plessis and Swartz, 1992; Everett et al.,
1994; Bowell, 2000; Geldenhuys et al., 2001; Younger et al., 2002).
This invited review will assess the options available for sulfate removal
from mine waters and discuss an economic and technical comparison made
between the available technologies

1.1 Sulfur hydrogeochemical cycle

Sulfur occurs in a number of oxidation states of which only three, the -2
(sulfide), 0 (native sulfur) and +6 (sulfate) are considered stable in nature,
with sulfate the dominant aqueous sulfur species found in most aerobic
geochemical systems (Chappelle, 1993). The dissolved concentration of
sulfate in mine drainage appears largely to be controlled by the solubility
of gypsum (~2.3 g/L CaSO4 assuming no other interactions). This solubil-
ity generally equates to ~ 500 mg/L Ca and 1800 mg/L sulfate (Figure 1).

The actual mechanisms involved

in sulfide oxidation are complex
and reviews have been published
elsewhere (Nordstrom, 1982;
Kleinnmann and Pacelli, 1991).The
most common source of sulfate re-
lease is oxidation of FeS2. Pyrite
and/or marcasite generate the
acidity of the mine waters and
simultaneously supply large quantities of Fe and sulfate, for example, one
tonne of ore with 1% pyritic sulfur can produce over 15 kg of ochre and 30
kg of sulfuric acid.
On weathering, sulfides can either release all sulfate or only a portion
and form secondary salts, such as halotrichite. These minerals are highly
soluble so can represent an instantaneous source of sulfate-rich water upon
dissolution and hydrolysis. (Cravotta, 1994; Alpers et al., 2003).
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 3

Sulfate reduction has been identified to occur in sediments (Fillipek and

Owen, 1980) natural wetlands and in mine waste (Kleinmann and Pacelli,
1991; Benner et al., 2002) and in natural gas reservoirs at low and high
temperatures (Trudinger et al., 1985). Over time in these locations metal
and sulfate concentrations have been observed to be lower and pH in-
creases to pH 7-8. The activities of sulfur species may be directly or indi-
rectly associated with the metabolism of microorganisms (Gould et al.,
1994). Indirect activities include the dissolution of the minerals under
acidic conditions that result from microbial metabolism, mineral precipita-
tion in anaerobic environments, adsorption to microbes and the formation
or disassociation of organometallic and arsine complexes. Direct metabolic
activity requires minerals as either soluble trace element within the cell
metabolic activity or to serve as specific oxidizing substrates, electron do-
nors/acceptors in the oxidation reduction activities.

1.2 Regulatory Control, Sulfate Corrosion and Toxicity

The accumulation of salts such as calcium sulfate in waters limits the
number of cycles of reuse of water on a mine site and creates environ-
mental problems if discharged (Sayre, 1988; DWAF, 1989). This is par-
ticularly a problem where mining is active in semi-arid to arid environ-
ments such as western USA, southern Africa, the Middle East, Central
Asia, parts of Australia and southern Europe. Demand on available water
in some places is such that much of the water is recycled. For example in
South Africa the average TDS of discharge water in the Rand Water Board
Area has risen from approximately 130 mg/L in 1935 to in excess of 1000
mg/L in 1980 (Heynike and McCulloch, 1982). Although not toxic in it-
self elevated sulfate (concentrations above 600 mg/L) in drinking water
has been observed to result in a purgation of the alimentary canal (WHO,
1996). Where high magnesium or sodium corresponds with high sulfate a
laxative and/or dehydration effect has also been reported (WHO, 1996).
Increasing concern has led to the introduction of recommended guide-
line values, rather than standards, for sulfate and TDS in groundwater and
effluent discharge (Table 1). These typically are based on USEPA or
WHO guidelines as to the maximum concentration of a particular chemical
constituent in drinking water or water utilized by livestock or for irrigation
and generally recommend no more than 500 mg/L sulfate (WHO,1996;
Sayre, 1988; DWAF, 1993; Flanagan, 1990; USEPA, 1999, 2002).
4 R.J.Bowell

Table 1: Recommended maximum Sulfate levels, in mg/l

Country Sulfate
USA effluent 500
Canada effluent 1500
EU guide limit 1000
DWAF effluent (South Africa) 600
Australia 1000
WHO guideline for drinking water 250

References: WHO, 1996; Sayre, 1988; DWAF, 1993; Flanagan, 1990;

USEPA, 1999, 2002

2. Treatment options and selection

Information on the treatment options presented was obtained from pub-
lished and unpublished sources from Europe, South Africa, Australia, and
the USA. The various approaches are discussed under two broad cater-
ogories; Membrane removal and Sulfur precipitation. A tabulated sum-
mary of each of the processes is presented in the appendix and are based
on previous compilations by Bowell (2000) and Lorax (2003).

3. Membrane removal of sulfate

Membrane removal of sulfate utilizes three possible methods; reverse
osmosis, electrical dialysis and filtration.

3.1 Reverse Osmosis

This process replies on a semi-permeable membrane which separates a
strong solution and a dilute solution. The greater the concentration differ-
ential across the membrane, the higher the tendency for water to permeate
to the concentrated solution. This hydraulic force is the osmotic pressure
of the system. In reverse osmosis an external hydraulic pressure is applied
to the saline brine thus forcing water through the membrane against os-
motic pressure.
In the case of brine concentration of ~ 30 g/L osmotic pressure is around
25 bars. Where water is low in calcium (<100 mg/L) and sulfate (<700
mg/L), conventional reverse osmosis can be used, although at higher con-
centrations scaling will occur. Modified processes have been proposed in-
cluding seeded reverse osmosis (SRO) (Harries, 1985) and special reverse
osmosis has been developed to treat mine waters (SPARRO) in South Af-
rica (Chamber of Mines Organization, 1988).
Seeded Reverse Osmosis (SRO) actively promotes precipitation of
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 5

CaSO4 prior to membrane treatment, reducing the corrosion of the mem-

brane walls and fouling by salt precipitation. This pre-treatment method
involves a suspension of seed crystals introduced into the effluent via re-
cycling of waste slurry. This crystal slurry is approximately 10% solids
and a tubular reverse osmosis (TRO) system is required because hollow fi-
bre systems are unsuitable because of fouling. A number of disadvantages
exist with SRO despite high salt and water recovery and reduced costs.
Energy consumption is high; there is poor control of CaSO4 seed and seed
solution controls.
Redevelopment of the SRO process led to the patent of the Slurry Pre-
cipitation and Recycle Reverse Osmosis (SPARRO) process (Pulles et al.,
1992). Extensive pilot plant test work has been undertaken and included
the precipitation of metals by increasing effluent pH to 10 as a pre-
treatment step. This is followed by cooling, filtration and readjustment of
pH 5-6 for protection of the membrane process. A pilot plant for SRO has
been operated in South Africa for 5000 hours and had a 96% water recov-
ery reducing sulfate from approximately 6600 mg/L to 150 mg/L A water
recovery of over 95% has been demonstrated by pilot studies. A problem
observed during pilot operation of the process was difficulty in maintain-
ing a flux rate of 550 L/m2/d due to fouling of the membrane, most likely
due to suspended SiO2 particles.
Laboratory testwork for a lower pressure RO circuit have been pub-
lished (Everett et al., 1994). At pressures of ~ 600 Kpa, at least 80% re-
covery of water was achieved. Several pre-treatment steps have been pro-
posed in order to extend membrane life. This includes chlorination to
remove bacteria, water softening to buffer pH and ion exchange to reduce
salt loading.

3.2 Electrodialysis and Electrodialysis Reversal (EDR)

This process uses direct electrical current across a stack of alternating
cation and anion selective membranes. In the effluent, anions are attracted
to the anode but cannot pass through anion-impermeable membranes and
are thus concentrated. Cations move in the opposite direction and are im-
peded by cation-impermeable/anion-permeable barriers. The initial con-
tainer has thus been deleted of salts and the cleaned water can be extracted.
By the use of current reversal the process is greatly improved. The anode
and cathode can be periodically changed as can the effluent and clean wa-
ter channels. This reduces potential for membrane fouling and facilitates
regeneration of the membrane by self-cleaning. A major advantage of EDR
over other RO techniques is that the system is not sensitive to effluent
temperature or pH. Capital costs are reduced as are working costs due to
lower working pressures. However CaSO4 scaling can occur due to inade-
6 R.J.Bowell

quate pre-treatment. A pilot plant study at Beatrix gold mine in South Af-
rica achieved a recovery of 80% salt ad recycled 84% water. The water has
high Fe, Mn, Na and Cl as well as sulfate (Juby and Pulles, 1990).

3.3 Filtration
Filtration is probably one of the more effective means by which sus-
pended particles can be separated from fluids (Buchanan, 1987). Mine wa-
ters are characterized by high suspended solid loadings which can be effi-
ciently removed by settlement of coarse material and filtration of the
majority of particles, most of which are less than 30 μm. Any filtration
process requires pre-treatment where coarse particles dominate suspended
load high fluid flow is present and consequently rapid rates of thickening
and filtration.
Common techniques of filtration involve the use of polyelectrolytes or
metal salts to act as a precipitating agent or target for flocculation. Physical
rather than chemical techniques are also available including screening,
freezing or thawing, elutriation and irradiation. A number of filtration op-
tions are available from deep bed filters which can be used to clarify fluids
to high purity, slow or continuous (or rapid) sand filters can be used to re-
duce turbidity and TDS.
Ultra-fine slurry particles can also be filtered using vacuum filtration
and electrolysis (Bollinger, 1984). Electrofiltration is particularly suited to
ultra-fine or colloidal particles (50% <2 μm). A well dispersed slurry is
placed in an active electric field results in migration of particles towards
the anode since they have a net negative charge. As most slurries are at
high pH the net negative charge is due to surface pH exceeding pHpzc im-
parting a negative charge on mineral surfaces (Ericksson, 1988). At the
anode a densely packed precipitate or cake is formed with a low water con-
tent. Through electro-osmosis the cake can be further dewatered, increas-
ing water recovery. The cathode essentially acts as a mechanical filter in
which a vacuum is created on the filtrate side forming a thin precipitate or
cake on the filter cloth. This acts as a trap for ultra-fine particles. Electro-
osmotic pressure operates in conjunction with the vacuum enabling the
production of a clear filtrate at higher rates than conventional vacuum fil-
ters (Bollinger, 1984).

4. Sulfur precipitation approaches

The removal of sulfur from mine waters by precipitation may be ap-
proached in one of three ways:
1. Precipitation of gypsum or mirabilite, Na2SO4.10H2O through ion ex-
change removal, and saturation through converging of pure cation or anion
streams.
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 7

2. Precipitation of an insoluble sulfate salt through chemical saturation

and precipitation to produce gypsum, CaSO4.2H2O (limestone/lime), bar-
ite, BaSO4 (addition of barium salts) or ettringite,
Ca6Al2(SO4)3(OH)12.26H2O (addition of alum , gypsum and lime).
3. Precipitation of sulfur or sulfide through biologically mediated reduc-
tion. This has the benefit of potentially also removing metal sulfides that
can be economically recovered as ore material (such as with the Biotech
process installed at the Bisbee mining operation, Arizona).

4.1 Ion-Exchange
Ion exchange resins are materials which contain large polar exchange
groups held together by a three dimensional network (Helffrich, 1962).
The process is an exchange of ions or molecules between solid and liquid
with no substantial change to the solid structure. One of the targeted ions is
essentially removed from the liquid phase and attached to the solid struc-
ture in exchange for another ion (typically hydrogen or hydroxyl) thus
rendering the target ion immobile.
In the case of CaSO4 sulfate, being an anion, would typically be ex-
changed for hydroxyl on a positively charged resin (an anionic resin)
while calcium, a cation, would be exchanged for hydrogen and so be at-
tached to a negatively charged resin (a cationic resin). As with reverse
osmosis, scaling of CaSO4 is common in conventional circuits. To over-
come these problems a modified form of ion exchange has been developed
to treat Ca-sulfate waters (GYPCIX).
The GYPCIX process (Gussmann and Nagy, 1993; Robertson et al.,
1993) is a novel process based on ion exchange resins which uses low cost
reagents such as lime and sulfuric acid for resin regeneration (Figure 2).
The resins used have been designed so as to target calcium and sulfate so
as to reduce gypsum levels in effluent thereby reducing TDS and corrosion
potential. Additionally a pure gypsum product is the result of both cationic
and anionic exchange and can be sold commercially thus offsetting treat-
ment costs. The reactions occur by mechanisms such as:
Cation Regeneration
R=Ca2+ + H2SO4 = 2 R=H+ + CaSO4
Anion Regeneration
R=SO42- + Ca(OH)2 + CaSO4

Pilot plant results in South Africa suggest that fouling caused by gyp-
sum precipitation in conventional ion exchange circuits can be avoided us-
ing the GYPCIX process (Robertson et al., 1993). As the waste streams
can be combined the gypsum in the slurry can be settled and the super-
natant water recycled so improving water recovery.
8 R.J.Bowell

CATION REGENERATION

In a pilot plant the effluent of

RCa+H2SO4 RH2 +CaSO4

REGENERATED
RESIN

Berkeley Pit, Butte, Montana was treated

H2SO4

by the GYPCIX process and the results

CaSO4

FEEDWATER

CATION EXCHANGE LOADING

CaSO4 + RH2 RCa + H2SO4
are given in Table 2. The GYPCIX
ANION REGENERATION

process can be used to treat solutions

RSO4 + Ca(OH)2 R(OH)2 +CaSO4

REGENERATED
RESIN

with sulfate up to 2000 mg/L and

LIME

CaSO4

calcium up to 1000 mg/L. Thereafter

AIR

membrane filtration is required to

ANION EXCHANGE LOADING DEGASSING
H2SO4 + R(OH)2 R2SO4 + H2O TOWER PRODUCT WATER
85% OF FEED

Figure 2: THE GYP - CIX PROCESS

(after Everett et al, 1994)
remove salts.
A variant of the GYP CIX process was developed by Feng et al. (2000)
to co-remove metals and sulfate. In this approach mine water was com-
pletely oxidized with hydrogen peroxide as a pre-treatment and this was
followed by magnetite-seed precipitation to create Fe-Mn hydroxide pre-
cipitate. This was accomplished at pH > 5 to prevent formation of H2S gas.
Sodium sulfide addition followed this to form metal sulfides, which were
precipitated at pH > 8 through lime addition. The Fe-Mn hydroxide-oxide
fraction can be magnetically separated from the base metal sulfide fraction.

Table 2: Results of GYPCIX treatment on Berkeley Pit effluent

(Robertson et al., 1993)

UNLIMED LIMED GYPCIX

TDS, mg/l 10000 3000 350

pH 2.7 8.5 8.0

Ca, mg/l 490 600 50

Mg, mg/l 420 350 20

Na, mg/l 70 70 50

Fe, mg/l 1100 0.1 <0.1

Mn, mg/l 182 3.6 <0.1

Cu, mg/l 186 <0.1 <0.1

Zn, mg/l 550 <0.1 <0.1

Sulfate, mg/l 8000 1980 200

 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 9

4.2 Chemical Saturation and Sulfate Salt Precipitation

Sulfate barriers may be constructed by using an inorganic source which
will produce a low solubility sulfate phase. Both lime and barium salts
have been proposed. With all the precipitation mechanisms the addition of
a chemically inert large particle material, such as carbon, to the reactor
feed will facilitate better settling of the sulfate precipitate. This is because
the ultra fine precipitates will attach to the larger particles and will thus
settle quicker than isolated fine sulfate particles.

4.2.1 Formation of Gypsum

The removal of sulfate by addition of lime or limestone occurs through
saturation of CaSO4. The precipitation of insoluble gypsum (Ksp ~ 10-2.3)
usually occurs as a by-product of lime addition in response to buffering of
drainage pH rather than designed remediation of sulfate concentrations
(Tahija et al., 1990). The impact of lime treatment on Berkeley pit effluent
can be observed to have a pronounced influence on water quality but is
nowhere as significant as Ba-salts (Table 3). Similar methods have been
proposed for neutralizing acidic drainage with both limestone and dolomite
(Maree and DuPleiss, 1994). Recent schemes proposed have demonstrated
the ability of semi-passive limestone/lime process to reduce sulfate in mine
waters to < 1000 mg/L (Gelddenhuys et al., 2001).

Table 3: Treatment of Sulfate-rich effluents by Ba- and Ca-salt pre-

cipitation
Shown as % removal

pH Lime BaCO3 BaS Ba(OH)2

2.9 62.3 24.2 95.6 107.7
7.9 80.5 101.6 110.8 137
12 51 90.1 90.1 134

4.2.2 Formation of Barite

The removal of sulfate by barium carbonate was demonstrated to be ef-
fective over 30 years ago (Kun, 1972) by a process such as:
BaCO3 (s) + SO42- = BaSO4 (s) + CO32-
Kun (1972) recognized three main problems with the method on an in-
dustrial scale, the requirements for more soluble Ba in solution than is re-
quired stoichiometrically, long retention times and high cost of Ba. The
cost problem can be overcome by recycling Ba through roasting of barite
10 R.J.Bowell

to form BaS and then purging with CO2 to reform BaCO3 (Wilsenach,
1986). Further consideration of BaCO3 to remove sulfate has come from
two research programs in South Africa in the late eighties (Trusler et al.,
1988; Maree, 1989; Maree et al., 1989; Adlem, 1997). The processes pro-
posed by these workers are summarized in Figure 3.
Barium carbonate and lime would be added to the effluent to soften the
water and produce a precipitate. From experimental work it was found
that CaCO3 was necessary to act as a seed to encourage BaSO4 formation
from BaCO3 due to the insolubility of the latter (Ksp ~ 10-8). Calcium salts
were found to achieve a better removal than Na or Mg salts as the products
(CaCO3/CaSO4) were much less soluble. The slurry from the reactor is
then sent to a thickener where clean water can be decanted and recycled in
the mine or process operation. The thickened slurry is then filtered, dried
and treated to recycle barium and collect sulfur. In a modification of this
process a two-stage fluidized bed reactor system has been proposed al-
though this process has difficultly with high metal concentrations and
separation of fine CaCO3 and BaSO4 (Maree et al., 1989).

UNTREATED
WATER
As an alternative to BaCO3, BaS has
BaCO3
been proposed (Maree et al., 1989;
Bosman et al., 1991) as a greater
REACTOR
CONDITIONING

quantity of sulfate is recovered but not TREATED

as much gypsum is produced; acid

WATER

THICKENER

waters can be treated directed,

eliminating the need for a pre-
FILTER
DRYING TRANSPORT

Figure 3: PROCESS FLOW SHEET FOR BARIUM neutralization step; and sludge disposal
CARBONATE PROCESS (modified from WRC 203/1/90)
(essentially gypsum) is greatly reduced.
The presence of metals in solution and production of metal sulfides can
significantly reduce the recovery of Ba from the process. This can be
avoided by sulfuric acid leaching which will oxidize sulfides, possibly
catalyzed by bacteria. Economically valuable metals can then be recovered
(Maree et al., 1989; Bosman et al., 1991). Alternatively the gas H2S can be
formed and vented through a wet scrubbing circuit to recover sulfur. A
possible flow sheet for this process is given in Figure 4.
CO2 SULFUR
S - PRODUCTION

The use of Ba(OH)2 has also been

STAGE

Air + CO2
CO2 + H2S

proposed to treat more neutral waters

Ca(HCO3)2 CaCO3
FLOCCULANT SOLUTION TREATED
SUSPENSION
CaCO3 WATER
SULFATE - RICH PRECIPITATION
CaS

where metals have been largely

SOLUTION BUBBLE
MINE WATER
BaSO4 REACTOR BED
SETTLER

precipitated already as metal hydroxide

salts (Adlem et al., 1991). The process
Air CaCO3
BaSO4 (can be used to buffer
water pH in sulfate-rich
acidic mine waters)
WATER BaSO4 COAL
BaS
PRECIPITATION REDUCTION

eliminates the necessity for complex

water treatment associated with the
SULFATE REMOVAL H2S STRIPPING WATER SOFTENING

Figure 4: SCHEMATIC FLOW SHEET OF BaS PROCESS

(modified from Maree et al., 1989)
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 11

BaCO3 and BaS processes, does not require long retention times for reac-
tions (such as in the BaCO3 process), and does not require stripping of H2S
as in the BaS process.
As a consequence of all of these reactions direct buffering of protons
can occur or, alternatively alkalinity is generated for neutralizing free pro-
tons. Because of the cost of barium and its environmental toxicity it is ad-
vantageous to have a barium recovery plant to recycle barium salts.
All three processes can remove sulfate from solution from very high
levels to within regulatory standards. In the case of BaS and Ba(OH)2
acidic solutions can be treated directly, although in practice some lime
treatment is required for very acidic solutions to prevent metal hydroxide
precipitation on the surface of the barium salt. The process additionally
removes transition metals, Mg, NH3 and, to a limited extent, Na. Thus the
overall TDS is lowered as well as the concentration of deleterious ele-
ments. The Ba(OH)2 causes significant CaSO4 precipitation improving sul-
fate removal by up to 30%, but increasing the volume of sludge requiring
disposal. A major benefit of the process is that valuable by-products are
created, the sale of which can be used to offset treatment costs. In the
BaCO3 and BaS processes sulfur, metals and Ba-salts can be commercially
produced while NaHS is produced in the Ba(OH)2 process. Overall the
BaS process is perhaps the most attractive for treating mine waters.

4.2.3 Precipitation of Ettringite

Sulfate removal through the precipitation of Ettringite has been pro-
posed by Smit (1999) as the SAVMIN process. The three stages of the
process successively remove metals as hydroxides through lime addition
(to pH 12) as a pre-treatment step (Figure 5).
This is followed by removal of gypsum through seed crystallization and
in the third step aluminum hydroxide addition to form insoluble ettringite.
Finally, prior to discharge, CO2 is added to reduce pH and precipitate pure
CaCO3. The ettringite can either then be disposed of or dissolved in sulfu-
ric acid to recycle Al(OH)3. The resulting effluent can then be seeded
with gypsum to produce more gypsum precipitate. Trials at the Stillfon-
tein plant in South Africa successfully treated 500 m3 of water with a sul-
fate concentration of 800 mg/L to < 200 mg/L. A variant on the SAVMIN
process is the Cost Effective Sulfate Removal (CESR) process. In addition
to sulfate, metals can also be removed by this process more effectively.
Other contaminants, such as nitrate, are removed during the ettringite pre-
cipitation step.

4.3 Biological Sulfate Reduction

Under anoxic conditions sulfate may be removed from mine waters as
12 R.J.Bowell

Lime Al(OH)3 CO2

stable sulfide precipitates. Under

these conditions sulfide minerals
Feed Product
Metal Gypsum Ettringite Carbonation
precipitation precipitation
water water

Hydroxides Gypsum
remain stable and have low
Decomposition
Reagent

solubility (Arenesen et al., 1991).

Gypsum
Recycle

Flooded
Sulfuric
acid
underground mine
workings and open pits can be
Figure 5: SAVMIN process flowsheet (Smit, 1999)
anoxic, and as such provide a
suitable environment for the implementation of a sulfate reduction system.
Alternatively a dedicated reactor can be used or anaerobic conditions cre-
ated in a passive system, such as a constructed wetland. The presence of
sulfides (H2S odor) in many mine water discharges indicates that sulfate
reduction is already occurring in mine workings (Huang and Tahija, 1990).
The reduction of sulfate to hydrogen sulfide is brought about by special-
ised strictly anaerobic bacteria and is accomplished primarily by two gen-
era: Desulfovibrio (five species) and Desulfotomaculum (three species).
These organisms have a respiratory metabolism in which sulfates, sulfites
and/or other reducible sulfur compounds serve as the final electron accep-
tors, with the resulting production of hydrogen sulfide (Chappelle, 1993).
The organic substrates for these bacteria are generally short chain acids
such as lactic and pyruvic acid. In nature these substrates are provided
through fermentative activities of other anaerobic bacteria on more com-
plex organic substrates (Gould et al., 1994).
Due to the natural occurrence of the sulfate reducing bacteria, sulfate
reduction can be utilized in situ for the treatment of acid rock drainage
provided the correct conditions can be maintained to sustain bacterial ac-
tivity. Anaerobic conditions may be enhanced by sealing shafts, adits and
air vents. For conditions to be sufficiently anaerobic however, it is likely
that a significant depth of water will be required. A wide variety of organic
substrates have been investigated for this purpose including molasses,
sewage sludge, straw, newspaper, sawdust and manure. Other possibilities
are wastes from the chemicals industry such as short chain organic acids.
Sulfide precipitation, like hydroxide precipitation, is not just dependent on
availability of constituent ions but also on environmental parameters. At
low pH copper and iron sulfides can be readily precipitated over a wide pH
range but zinc, which forms sulfides in a similar way, has a much slower
rate of formation than copper or iron sulfides. Neutral or mildly alkaline
conditions tend to be the most favorable for sulfide formation.

4.3.1 In-situ reactor

Because sulfate reducing bacteria occur naturally there is a possibility of
treating high sulfate waters using an in situ process So far fixed bed reac-
tors and in-pit reactors have been utilized (Arnesen et al., 1991; Huang and
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 13

Tahija, 1990; Robins et al., 1997) but stirred reactors with a suspended
solid medium have also been proposed, with the aim of achieving higher
reduction rates through improved operating conditions and reactor utiliza-
tion.
Based on limnological-microbiological-geochemical studies a series of
three zones have been described for the Summer Camp Pit lake in Nevada
(Bowell, 2002) . The upper oxic zone is characterized by high levels of to-
tal epifluorescent algae and heterotrophic aerobic bacteria and has high
dissolved oxygen content and consequently total sulfur is dominated by
sulfate. This zone extends to a depth of 6 m, after which a transitional
zone develops, which is characterized by an increase in the presence of
heterotrophic anaerobic bacteria and decreasing levels of total epifluores-
cent algae and heterotrophic aerobic bacteria. Within this zone sulfates
(and thiosulfate) are being gradually reduced to sulfide. This zone grades
into a zone with little or no available dissolved oxygen and is dominated
by heterotrophic anaerobic bacteria. In this zone sulfate is being actively
reduced to sulfide and the potential exists for the precipitation of metal sul-
fides. The addition of raw potato and steer manure amended systems in-
creased sulfate reduction (Gannon et al., 1996). It was found that raw po-
tato-stimulated sulfate reduction was effective at low levels whereas steer
manure stimulated sulfate reduction by bacteria at mid-high levels. Using
2500 mg C/l raw potato, 80% sulfate reduction was achieved and addition-
ally soluble arsenic, after an initial increase, decreased to less than 1% of
initial values in some cells, presumably due to formation of arsenic sulfide.

4.3.2 Constructed Bioreactors

Anaerobic bioreactors use bacterial reduction of sulfate and iron to ac-
complish precipitation of metal sulfides (Figure 6).
BIOREACTOR CELL

PERFORATED PIPE
These reactions can only occur
STRAW
CONSTANT HEAD

in an anaerobic reactor as the

RESERVOIR
GEOFABRIC

EFFLUENT
DISCHARGE

bacteria cannot tolerate oxygen.

Effluent treatment therefore
SUBSTRATE
GEOFABRIC

requires a uniform rate and flow

GRAVEL BED

IMPERMEABLE
CONSTRUCTION MATERIAL

PERFORATED PIPE
through the substrate to be
PREPARED SOIL BED
VALVE

Figure 6: SCHEMATIC OF A TYPICAL UPFLOW

ANAEROBIC BIOREACTOR
effective. In an experimental
reactor the removal of sulfate was
found to be dependent on energy source, flow and time (Dill et al., 1994).
The biological reduction of sulfate from industrial effluents has been
shown to be an effective method with producer gas being a reliable energy
source (Maree et al., 1991; Du Preez et al., 1992). From studies by Dill et
al. (1994) a pilot plant produced a sulfate reduction rate of 1.7-2.2 g/l/d
14 R.J.Bowell

over 70 hours reducing effluent drainage sulfate from over 3000 mg/l to
less than 250 mg/l.

4.3.3 Constructed wetlands

In recent years constructed wetlands have become increasingly popular
in dealing with closed mines where little or no records are preserved and it
is not possible, in general, to prevent or contain the effluent (Machemer et
al., 1993; Younger et al., 2002). In the Pelenna valley a pilot scheme was
assessed for the effectiveness of wetlands in removing sulfate (Rees and
Bowell, 1999). The design discharge flow rate is 3 l/s at an average pH
5.6 with an average total iron content of 21.7 mg/l and sulfate content of
459 mg/l. Over the ten year period of monitoring on the pilot scheme sul-
fate has been observed to decrease. Sulfur isotope analysis of pore waters
from the scheme indicated that whilst sulfate reduction was occurring in
the beds, subsequent metal sulfide oxidation offset the treatment with the
net effect that in the long term sulfate attenuation would not occur (Rees
and Bowell, 1999).

4.3.4 Permeable Reactive Barriers

Permeable Reactive Barriers are a reactive zone in an aquifer created for
removal of contaminants by reaction with an introduced material. Al-
though no specific PRB has been developed for sulfate treatment, the re-
duction in high levels of sulfate in groundwater at an Ontario mine site has
been reported (Benner et al., 1999). In this system a mixture of gravel and
compost was used to create a biological sulfate reduction zone that precipi-
tated iron sulfide at a rate of sulfate removal of 14 mg/L/day over a 3-year
period. A major limitation on the application of such a system is the re-
quirement for stoichiometrically equivalent amounts of reduced metals to
sulfide ions in order to limit sulfur dispersion.

5. Evaluation of treatment options for sulfate removal

The criteria by which one of these options is chosen typically depend on
the ability of the process to adequately remove sufficient sulfate in a given
time such that discharge of the treated water meets all regulatory require-
ments. A further (as important) criterion is that of economics. In any min-
ing operation, the costs have to be balanced with the income. In a feasibil-
ity study, all the costs are assessed including the costs of environmental
protection. The approach with the design of any aspect of the mine is gen-
erally BATNEEC (Best Available Technology Not Entailing Excessive
Cost). The detail will be a function of the planning authority requirements
and the philosophy of the mining company.
The approach taken for water treatment generally involves assessment
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 15

of treatment alternatives and risk assessment for release of contaminants.

However, a mine is developed on the basis of a return on investment.
There is a limit as to the environmental costs beyond which the mine will
not go ahead. Although risk assessments are done, the methods of assess-
ment and monitoring cannot be considered as exhaustive and engineering
decisions are made at some stage of investigation, which are commensu-
rate with the level of investment in the mining project itself.
Whilst a direct evaluation of the differing merits of the different sulfate
removal options outlined in this review is difficult to compare, due to the
importance of site-specific requirements, some general comments can be
made. Membrane and ion-exchange type processes tend to have high capi-
tal and operating costs associated with them (Appendix). Many of the
processes also are susceptible to fouling of the membrane or column and
virtually all require some form of pre-treatment. In terms of mine water
treatment GYPCIX and SPARRO appear to be the most suitable for treat-
ment of high sulfate waters. However, in an economic evaluation of re-
quirements for the treatment of water at the Grootelvei mine in South Af-
rica (Schoeman and Steyn, 2001) operating costs and associated savings
were such that EDR and GYPCIX processes were more favourable than
RO methods (Table 4).

Table 4: Calculated costs for the treatment of Grootvlei mine water

(from data published by Schoeman & Steyn, 2001)

Reverse Osmosis EDR GYPCIX

Unit cost, US$/m3 0.88 0.48 0.60
Annual operating costs,
21.9 9 9.6
$M
Brine disposal,
0.82 1.9 2.5
US$0.19/m3
Annual saving (drinking
5.5 4.2 3.5
water)
Total annual operating
17.2 6.7 8.6
costs, US$M1

1
Calculated costs for a 80,000 m3/day plant

Chemical precipitation processes, although not widely demonstrated,

show some potential for application. However they all require high quan-
tities of relatively pure chemicals for successful operation and produce
high volumes of waste. Of the processes reviewed the limestone/lime and
SAVMIN processes are the most applicable to mine waters. Where cheap
16 R.J.Bowell

sources of Ba-salts can be procured the Ba-precipitation processes could

also be highly effective in reducing sulfate in mine effluents.
In recent years the majority of research related to sulfate removal has
focused on biological sulfate reduction and currently these are the most
widely applied methods (after addition of lime/limestone) for sulfate re-
moval. The advantage of processes such as the Biotech method in produc-
ing a commercially value by-product (metal sulfide concentrate), the low
volume of resulting solid waste, and the relative inexpensive approach,
make bioreactors the most favourable approach currently in sulfate re-
moval from mine waters.
Where sulfate levels are very low there is no requirement for sulfate
treatment. Below sulfate levels of approximately 500-1500 mg/l, passive
measures can be utilized, while at higher levels (1500-10000 mg/l) a wide
range of options are available including Reverse Osmosis, SAVMIN,
GYPCIX, filtration, desalination, bioreactors and salt precipitation. At
very high levels of sulfate (> 10000 mg/l) co-precipitation is probably
most suitable, options for which include SAVMIN, CESR and possibly
bioreactors.

6. Conclusions
Sulfate removal from mine waters is still considered to be of secondary
importance compared to removal of metals and acidity. As such it is com-
paratively less understood and few case studies exist for evaluation.
Future trends in regulatory practice may promote more research and ap-
plication of sulfate removal technologies as desalination of mine water ef-
fluent becomes more important.
Various options exist for sulfate removal. Chemical precipitation is
generally the least costly, but produces high volumes of waste by-products.
The SAVMIN process, although expensive in terms of proposed operating
and capital costs, offers the most efficient treatment of high sulfate waters
by precipitation methods.
Membrane and ion exchange processes, with the exception of SPARRO
and GYPCIX, are not suitable for mine water treatment unless extensive
pre-treatment is applied.
Biological Sulfate Reduction offers perhaps the most versatile and
widely applicable approach to sulfate removal from mine waters and has
the benefit of being able to couple sulfate and metal removal.
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 17

Acknowledgements
The author would like to acknowledge the assistance of the laboratories
which have been involved in the analysis associated with this work and the
clients who approved publication of data.
This work has particularly benefited from discussions with Dr Janee
Maree of CSIR Water Technology Division in Pretoria, South Africa and
his time in sharing his knowledge is greatly appreciated. In addition the
sections on biological sulfate reduction have been influenced by discussion
and publications provided by Ben Rees of SRK and Dr James Gannon of
the Montana State University. Dr Adam Jarvis is thanked for his review of
the initial draft of this paper.

References
AWWA, 1999 Water Quality and Treatment. 5th Edition, American Wa-
ter Works Association. McGraw Hill Inc., New York.
Adlem, C.J.L., Maree, J.P., and Pleiss, P.Du. 1991 Treatment of sul-
fate-rich mining effluents with the Barium Hydroxide process and recov-
ery of valuable by-products. In: 4th International Mine Water association
Congress, Ljubljana (Slovenia)-Portschach (Austria), September 1991,
211-221.
Adlem, C.J.L. 1997 Treatment of sulfate-rich effluents with the barium
sulfide process. MSc thesis. University of Pretoria, South Africa.
Alpers, C.A., Nordstrom, D.K. ad Spitzley, J. 2003 Extreme acid mine
drainage from a pyritic massive sulfide deposit: The Iron Mountain end-
member. In: Environmental Aspects of Mine Wastes, editors: J.L. Jambor,
D.W. Blowes and A.I.M Ritchie. Mineralogical Association of Canada
Short Course Series No. 31, 407-430.
Arnesen, R.T. et al. (1991) Monitoring water quality during filling of
the Lrkken mine: Role of sulfate reducing bacteria in metals removal. 2nd
Int. Conf. Abatement Acid Drainage, Montréal, Canada, 16-18/9/91.
Benner, S.G., Blowes, D.W., Gould, W.D., Herbert, R.B., and Placek,
C.J. 1999 Geochemistry of a permeable reactive barrier for metals and acid
mine drainage. Environmental Science and Technology, 33, 2793-2799.
Bollinger, J.M. 1984 New filtration concepts using Electrodeposition.
Chem.Eng., May 1984: 327-374.
Bosman, D.J. 1985 Productive use of water in mining. SA Water bulle-
tin. February 1985.
Bosman, D.J., Clayton,J.A., Maree, J.P. and Adlem, C.J.L. 1991 Re-
moval of sulfate from mine water. In: Proceedings of the ARD in Pyrite
Environments, Lisbon, Portugal, 16-19 September 1990. 211-221
Bowell, R.J. 2000 Sulfate and salt minerals: the problem of treating
mine waste. Mining and Environmental Management, 11-14.
18 R.J.Bowell

Bowell, R.J. 2002 Hydrogeochemical dynamics of Pit Lakes. In:

Younger, P.L. and Robins, N. (eds) Mine Water Hydrogeology and Geo-
chemistry. Geological Society of London Special Publication, No.198,
159-186.
Buchanan, J. 1987 Basics of filtration. Water Sewage & Effluent. Sep-
tember 1987; 35-36.
Chamber of Mines Research Organization 1988 New desalination pro-
gram on stream. R&D News CM. October 1988.
Chappelle, F.M. 1993 Groundwater Microbiology and Geochemistry.
Prentice-Hall.
Cravotta, C.A. III 1994 Secondary iron-sulfate minerals as sources of
sulfate and acidity. In: Environmental Geochemistry of Sulfide Oxidation
(eds: C.N. Alpers and D.W. Blowes). Am.Chem. Soc. Sym. Series, 550.
1994. 345-364.
Department of Water Affairs and Forestry, South Africa 1993 South Af-
rican Water Quality Guidelines 4 vols. DWAF, SA.
Dill, S., Du Preez, L., Graff, M. and Maree, J. 1994 Biological sulfate
removal from acid mine drainage utilizing producer gas as carbon and en-
ergy source. 5th International Mine Water Congress, September 1994: 631-
641.
Du Pleiss, G.H. and Swartz, J.A. 1992 Tubular Reverse Osmosis treat-
ment of Secunda mine water: a pilot plant investigation. Water Science
Technology, 25, 193-201.
Du Preez, L.A., Maree, J.P., and Jackson-Moss, C.A. 1992 Biological
sulfate removal from industrial effluents. Environ. Tech., 13: 875-882.
Eckenfelder, W W 1989 Industrial Water Pollution Control. McGraw
Hill, 400pp
Eriksson, P. 1988 Nanofiltration extends the range of membrane filtra-
tion. Environmental Progress, 7: 58-62.
Everrtt, D.J., Du Plessis, J. and Gussman, H.W. (1994) The removal of
salt from underground mine waters. Mining Env. Management, March
1994: 12-14.
Feng, D., Aldrich, C., and Tan, H. 2000. Treatment of acid mine water
by use of metal precipitation and ion exchange. Minerals Engineering,
13(6), 623-642.
Fillipek, L.H. and Owen, R.M. 1980 Early Digenesis of organic carbon
and sulfur in outer shelf sediments from the Gulf of Mexico. Am. J.Sci.,
280: 1097-1112.
Flanagan, P.J. 1990 Parameters of water quality: Interpretation and
standards. Environmental Resources Unit. 160pp.
Gannon, J., Wielinga, B., Moore, J.M., Policastro, P., McAdoo, D and
Meikle T. 1996 Field Investigations of the Sulfate reducing potential in the
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 19

Summer Camp Pit lake. Bitterroot consultants. 1996. 12pp.

Geldenhuys, A.J., Maree, J.P., de Beer, M. and Hlabela, P. 2001 An in-
tegrated limestone/lime process for partial sulfate removal. Proceedings of
Environmentally Responsible Mining in South Africa, September 2001.
CSIR, Pretoria, South Africa.
Gould, W.D., Bechard, G. and Lortie, L. 1994 The nature and role of
microorganisms in mine drainage. In: Environmental Geochemistry of
Mine Waste, (eds: J.L.Jambor and D.W. Blowes). 185-199, MAC.
Gussman, H.W. and Nagy, P. 1993 The development and operation of
the GYPCIX pilot plant, JCI MPRL, Johannesburg, SA, July 1993.
Harries, R.C. 1985 A field trial of Seeded Reverse Osmosis for desalina-
tion of a scaling type mine water. Desalination, 56: 227-236.
Haynike, J.J.C. and McCullough, S.F. 1982 The impact on an Urban and
Industrial complex of mineral pollution in the water supply. Wat. Sci.
Tech., 14: 381-395
Helfferich, F. 1962 Ion Exchange. McGraw-Hill, San Francisco: 624 pp.
Huang, H-H. and Tahija, D., 1990 On the nature of Berkeley Pit Water.
Proceedings Mineral and Hazardous Waste Processing Symposium. Butte
Montana 10/01/90.
Juby, G.J.G. 1989 Membrane desalination of mine water. WISA Con-
ference. March 1989.
Juby, G.J.G. and Pulles, W. 1990 Evaluation of Electrodialysis reversal
for desalination of brackish mine water. WRC Report 179/1/90.
Kleinmann, R.L.P. and Pacelli, R.R. 1991 Biogeochemistry of acid
mine drainage. Mining. Eng., 33, 300-6.
Kun, L.E. 1972 A report on the reduction of the sulfate content of acid
mine drainage by precipitation with barium carbonate. Anglo American
Research Laboratories. D/3/W/1.
Lorax 2003 A review of sulfate treatment. Report to INAP by Lorax
consultants. Electronic document, INAP website.
Machember, S.D., Reynolds, J.S., Laudon, L.S., and Wildeman, T.R.
1993 Balance of sulfur in a constructed wetland built to treat acid mine
drainage, Idaho Springs, Colorado. Appl. Geochem., 8: 587-603.
Marree, J.P. 1989 Sulfate removal from Industrial effluents. PhD thesis.
Univ. Orange Free State.
Maree, J.P., Bosman, D.J. and Jenkins, G.R. 1989 Chemical removal of
sulfate, calcium and metals from mining and power station effluents. Wa-
ter Sewage and Effluent, 9: 10-25.
Marree, J.P. and Du Plessis, P. 1994 Neutralization of acid mine water
with calcium carbonate. Wat.Sci.Tech., 29: 285-296.
Maree, J.P., Hulse, G., Dods, D. and Schutte, C.E. 1991 Pilot plant
studies on biological sulfate removal. Water Pollution Research and Con-
20 R.J.Bowell

trol, part 3. CSIR.

Nordstrom D.K. 1982 Aqueous pyrite oxidation and the consequent
formation of secondary minerals. In: Acid Sulfate Weathering, 37-56. Soil
Sci. Soc. Am.
Pulles, W., Juby, G.J.G., and Busby, R.W. 1992 Development of the
Slurry Precipitation and Recycle Reverse Osmosis Technology for desali-
nating mine waters. Water, Sci. Tech., 25. 25pp.
Rees, S.B. and Bowell, R.J. 1999 Stable Isotopic Modeling of the Lon-
gevity of Treatment Processes Operating in a Constructed Wetland for the
Amelioration of Acid Mine Drainage. In: IMWA 20th Mine Water Confer-
ence, Seville, September 1999, 585-590
Robins, R.G. et al.1997 Chemical, physical and biological interaction at
the Berkerley Pit, Butte, Montana. Tailings and Mine Waste `97. 529-541.
Balkema, Rotterdam.
Robertson, A.M., Everett, D., Du Plessis, P. 1993 Sulfate removal by
GYPCIX process following lime treatment. In: SUPERFUND XIV Con-
ference, 30/11-2/12/93, Washington USA.
Sayre, I.M. 1988 International Drinking Water Standards. Journal of
AWWA, January 1988.
Schoeman J.J. and Steyn, A. 2001 Investigations into alternative water
treatment technologies for the treatment of underground mine water dis-
charged by Grootelvei Proprietary Ltd into the Blesbokspruit, South Af-
rica. Desalination, 133, 13-30.
Smit, J.P. 1999 The purification of polluted mine water. Proceedings of
the International Symposium on Mine Water & Environment for the 21st
Century, Seville, Spain.
Tahija ,D., Huang, H-H., Flores, C., and Shi, Y. 1990 On the treatment
of Berkerley Pit Water. Proceedings Mineral and Hazardous Waste Proc-
essing Symposium. Butte Montana 10/01/90.
Trudinger, A., Chambers, L.A., and Smith, J.W. 1985 Low temperature
sulfate reduction: biological versus abiological. Can.J.Earth Sci., 22, 1910-
1918.
Trusler, G.E., Edwards, R.I., Brouckaert, C.J., and Buckley, C.A. 1988
The chemical removal of sulfates. Proceedings of the 5th National Meeting
of the South African Institute of Chemical Engineers, Pretoria, W3-0-W3-
11.
USEPA, 1999 National Recommended Water Quality Criteria Correc-
tion. EPA 822-Z-99-001, Washington DC.
USEPA, 2002 Edition of Drinking Water Standards and Health Adviso-
ries. EPA 822-R-02-038, Washington DC.
WHO 1996 Guidelines for drinking water quality. Geneva (2 vols.), 2nd
edition.
 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 21

Wilsenbach, I., 1986 Cost estimate for barium sulfate reduction. Inter-
nal report of the Division of Water Technology, CSIR, 620/2616/6.
Younger, P.L., Banwart, S.A., and Hedin, R.S. 2002 Mine water: Hy-
drology, Pollution, Remediation. Kluwer Oordecht.
 Table A.1
Summary of case studies on treatment processes using membranes and ion-exchange (after Bowell, 2000; Lorax, 2003)
RO SPARRO EDR GYP-CIX Filtration
Pretreatment yes yes yes no Yes
Feed water sulfate SO4: any SO4: any SO4: any SO4: any SO4: any
limits
Sulfate removal SO4:>99% SO4:>99% SO4:>95% SO4:>95% SO4:>95%
Brine production yes yes yes yes yes
Sludge production low low low low-moderate moderate-high
Monitoring low-moderate low-moderate low-moderate low moderate-high
Maintenance high high high moderate high
1 1 1
Capital cost £0.46 M £0.48 M 1 £0.39 M £0.22 M £ 0.28 M1
per 103 m3 / day per 103 m3 / day per 103 m3 / day per 103 m3 / day per 103 m3 / day
Operating costs £ 0.49 / m3 £ 0.17 / m3 £ 0.27 / m3 £ 0.33 / m3 £ 0.15 / m3
Advantages - drinking water - drinking water quality - drinking water - drinking water - drinking water
quality - improved membrane life quality quality quality
Disadvantages - scaling problems - short membrane life - scaling problems - sludge production - scaling problems
- short membrane life - short membrane life - short membrane life

Improvements - not suitable for - membrane life - not suitable for - sludge recycling
scaling waters scaling waters - metal recovery

1 Conversion used: South African R 15 = £1 = USD1.78)

 A REVIEW OF SULFATE REMOVAL OPTIONS FOR MINE WATERS 23

Table A.2
Summary of case studies on treatment processes with mineral precipitation
Limestone / Lime BaS SAVMIN CESR
Pretreatment no no no no
Feed water sulfate lim- SO4: any SO4: any SO4: any SO4: any / L
its
Sulfate removal 50 % >98% >90% >95%
Brine production no no no no
Sludge production low-moderate low-moderate moderate-high high-very high
Monitoring moderate-high high high high
Maintenance low low low low
Capital cost 2 £ 0.13 M per 103 m3 / day £ 0.25 M per 103 m3 / day £ 0.18 M per 103 m3 / day £0.22 M per 103 m3 / day
(ΔSO4: 2,000mg / L)
Operating costs1 £ 0.34 / m3 £ 0.22 / m3 £ 0.12 / m3 £ 0.44 / m3
(SO4: 2,000mg / L) (SO4: 1,500mg / L)
Advantages - also trace metal removal - low levels of sulphate - low levels of sulphate - low levels of sulphate
- very cheap - recycling of expensive BaS - recycling of ettringite - also trace metal removal
- also trace metal removal
Disadvantages - limited sulphate removal - little trace metal removal - production of sludges - production of sludges
- production of sludges - production of sludges
Improvements - recycling of sludges - recycling of sludges - recycling of sludges - recycling of sludges
 24 R.J.Bowell
View publication stats

Table A.3
Summary of case studies on treatment processes using biological sulphate removal
Bioreactor Constructed Wetland Alk. Producing systems Permeable barrier
Pre-treatment yes yes yes no
Feed water sulfate SO4: any SO4: <2000mg / L SO4: <2000mg / L SO4: <2000mg / L
limits
Sulfate removal >90 % >50% >50% >80%
Sludge production low-moderate moderate-high moderate-high moderate
Monitoring moderate-high low low low
Maintenance moderate low low low
SO4 reduction rate 5-30g / L, day 0.03-0.2mg / L, day 0-0.05mg / L,day (low) <1-20mg / L,day
Capital cost £ 0.18 M per 103 m3 / day £0.06 M per 103 m3 / day £0.08 M per 103 m3 / day £45,000
(ΔSO4: 2,500mg / L)
Operating costs £0.17 / m3 £0.05 / m3 £0.06 / m3 £16,850 / yr
(ΔSO4: 2,500mg / L)
Advantages - also trace metal recovery - also trace metal removal - gypsum precipitation - passive treatment
- recycling of H2S and CO2 - passive treatment - also (trace) metal removal - also trace metal removal
- low maintenance
Disadvantages - cost of C + energy source - little sulphate reduction - sludge disposal - long-term performance
- production of sludge - sludge disposal - limited time life unknown
- limited time life - prone to scaling
Improvements - recycling of sludge - specific design required - specific design required - alternative reactive media
- cheap C + energy source